Last week, search giant Google’s chief executive officer, Sundar Pichai paid the Pentagon a visit during his trip to Washington in order to reduce the tensions in the premises of the tech giant, post an employee’s outrageous move to severe the defense contract to analyze drone video.
Some people from the company, who agreed to speak on conditions of anonymity said that Pichai met a group of civilians and military leaders who were from the office of the Undersecretary of Defense for Intelligence and the Defense Department directorate respectively, who oversaw the artificial-intelligence drone system that is known as “Project Maven”.
Google has worked with the Defense Department during the development of Project Maven, which consequently makes use of AI technology to tag cars, buildings and other objects in the videos recorded by drones that fly over conflict zones automatically. After a series of events, in the month of June, Google announced that it has no plans of renewing its contract with the Defense Department, due to the increase in the number of employees who criticized the work as “actually helping the military track and kill with greater efficiency”.
A Defense Department spokesperson added- “We do not wish to comment on the details of private meetings. Department leaders routinely meet industry partners with the goal of discussing innovative technologies. These meetings help in supporting the continuing dialogues that aim at solving future the technology challenges.”
Pichai’s visits highlight the fact that Google has been trying to retain its workers of Silicon Valley who were triggered by the moral implications of the developments in the warfare technology in addition to trying to get Washington’s lucrative military contracts. Previously, Google CEO, Pichai, in a blog post, said Google has and it will continue to work with the defense leaders to focus on “cyber-security, training, military recruitment, veterans’ health care, and search and rescue.” In addition to this, Google has also bid for one of Pentagon’s most lucrative cloud-computing contracts.
Google also faced backlash this year due to its involvement in Project Maven because of which more than 3,000 workers addressed an open letter to Pichai saying “Google should not be in the business of war.” Critics added that AI technology could result in much more devastating drone strikes and might result in concerns such as “killer robots” and other lethally autonomous machines.
A source said that military officials were not happy with Google’s change-of-heart over Project Maven which was actually their first big AI collaboration with the Pentagon. The officials felt that Google should have communicated in a better way and informed them about the technologies that would help them keep service members out of danger.
Apart from this, Google’s decision to terminate the collaboration with Project Maven has brought in a lot of flak from congressional lawmakers, especially Republicans.
Bob Work, the former deputy secretary of defense who helped launch Project Maven last year said-“Without a doubt, this has caused a lot of consternation inside the DOD. Google created a big moral hazard for itself by telling that it doesn’t wish to use any of its AI technology to kill humans. But Google didn’t mention anything about the lives that could be saved using the same technology.”
It is expected that Pichai has been asked to hold a press meet (which hasn’t been scheduled yet) on the matter wherein he will be asked to testify the matter. In September, GOP Sen. Tom Cotton, R-Ark., blamed Google at a hearing stating that the executives of the company declined to appear because Google couldn’t help aid the government on AI tools “which had been designed not just to protect our troops, and help them fight in our country’s wars, but to protect civilians as well.”
Cotton and his peers, during a September hearing, reported that Google is “working in order to develop a new search engine which will satisfy the Chinese Communist Party’s censorship standards.”
Cotton and three other GOP lawmakers criticized Google for helping aid Chinese companies as well as withdrawing from partnerships with the DOD.
Project Maven was the first advanced AI project which was being used for an operational combat zone, that inspired a broader debate over the potential dangers of deploying a powerful machine-learning technology and “weaponizing AI” into a theatre of war.
The AI for Project Maven was officially known as the Algorithmic Warfare Cross-Functional Team, which relies on the same type of “computer vision” techniques that are currently being used by consumers for image-recognition software and this includes Google as well.
During the month of June, Google announced that it will not be extending its 18-month DOD subcontract deal which is due in March for expiration. Google unveiled a set of AI ethical principles, that included an internal ban on development of AI which might potentially be used in weapons or which could cause overall harm. They were general guidelines and did not mention details regarding how these tules can be enforced practically.
Gregory Allen, an adjunct fellow for the Washington think tank Center for a New American Security, said- “Google’s sudden plans of retorting back from the collaboration was an embarrassment and this was a result of communications fiasco that was for an important DOD initiative. This also put the company’s very promising prior courtship in peril.”
Allen said “Google’s credibility as a company to collaborate with vital national security work has been hurt very badly due to the Maven pullout.”